Room for one more eddy in a stream

The notion of turbulence evokes thoughts of erratic and disordered, and maybe violent and unpredictable activity. Macaulay spoke of turbulent times and Milton of turbulent minds. Engineers with preoccupations with subjects such as metrology, aerospace, combustion engineering and a spectrum of other engineering disciplines have an apparently more precise basis for the rationalisation and description of their interest.

But books such as this one by Stephen Bailey Pope indicate that the science and engineering of physical turbulence has many ill-defined elements of the sort that prevent the adequate description of human turbulence.

At a fundamental level, the turbulent flow is governed by complex interparticle interaction. Molecules and their interactions are not as unpredictable as people and their interactions. Nevertheless, some features of turbulent flows do have humanoid and social characteristics.

There is no doubting the practical importance of understanding the origins and nature of turbulent flows; they occur widely in nature and as a result of human endeavour. The subject has been, and continues to be, extensively studied and reported. This is a graduate-level textbook based on a graduate course, and it will be useful for that purpose. Students studying turbulence at that level will now have another textbook to consult.

Turbulent Flows covers the conventional, expected topics in two parts. The first provides an introduction that takes the reader through such areas as Navier-Stokes equations, statistical representations, shear and wall-boundary flows and turbulence spectra.

The second part of the book describes a variety of well-practised modelling and simulation procedures for describing turbulence: included are all the models that would be familiar to those who are involved in the subject. This part occupies about half the volume and covers direct numerical simulation, turbulent viscosity models, Reynolds-stress models, probability density function approaches and the large eddy-simulation approximations.

In all, the content of Turbulent Flows is rather standard and is similar to that of the recent An Introduction to Turbulent Flow by Jean Mathieu and Julian Scott. The two books seek to provide introductory textbooks for similar audiences. But in spite of the common aim and overall content, the two are rather different in style and detail.

The Mathieu and Scott text has a much lighter feel and takes a less rigorous approach to provide a broad picture; the authors are aware of this and will, I understand, publish a more detailed account of the subject. Pope's book has more detail at the expense of stressing the broader facets. Both books are good in their own ways and are in a sense rather complementary and provide a nice introduction to the subject.

At the level of detail, Turbulent Flows has a few limitations for students. There are many exercises to test the reader's appreciation of the subject but no solutions. There are references, but they are not listed by topic; searching is therefore not straightforward. Little is said of experimental techniques.

Turbulent Flows is a useful addition to the teaching archives. There are other books on a similar theme, but we have not reached the stage where, as was said some years ago of books on thermodynamics: "We need another book on thermodynamics like we need another hotel in Miami!" The book does not cover the whole ambit of turbulent flows, an area that is difficult to describe, but what it does it does well. One hopes it will be widely read.

Brian Briscoe is professor of interface engineering, Imperial College, London.